sign in sign up
<<
Home , Apical ectodermal ridge, TBX5 (gene)

Developmental Biology 300 (2006) 623–634 www.elsevier.com/locate/ydbio

Blimp-1 is an essential component of the genetic program controlling development of the pectoral bud ⁎ Ban Chuan Lee, Sudipto Roy

Institute of Molecular and Cell Biology, Proteos, 61 Biopolis Drive, 117688, Singapore Received for publication 7 March 2006; revised 13 June 2006; accepted 26 July 2006 Available online 4 August 2006

Abstract

Formation of paired limbs in vertebrate embryos has long been a particularly useful paradigm for the study of pattern formation. Here, we show that Blimp-1, a SET domain and zinc -containing transcriptional factor, plays an important role in the development of the pectoral fins of the zebrafish structures that are homologous to of amniotes. The blimp-1 is expressed dynamically in the as well as the ectodermal cells of the early fin bud, and later, in the cells of the (AER) of the outgrowing fin. Consistent with this expression profile, loss of Blimp-1 activity severely impairs fin outgrowth and patterning. We present evidence that blimp-1 functions downstream of and fgf24 and therefore is not required for the initial specification of the fin bud primordia. Subsequently, however, its function is necessary for the induction of and in the mesenchyme. In addition, Blimp-1 activity is absolutely critical for the proper induction of in the and establishment of the AER. Taken together, these results identify an additional layer of control in the genetic pathway that operates in the developing limb and provides novel insights into regulatory mechanisms that organize its pattern. © 2006 Elsevier Inc. All rights reserved.

Keywords: Blimp-1; ; Zebrafish; Pectoral fin; tbx5; fgf24; fgf10; Apical ectodermal ridge; Zone of polarizing activity; Sonic hedgehog

Introduction et al., 2001; Ng et al., 2002). It should be noted, however, that recent genetic studies with the mouse do not corroborate such Vertebrate limbs begin their development as localized an early role for Fgf8 signaling from the IM in limb bud protrusions called limb buds in the lateral plate initiation (Boulet et al., 2004; Perantoni et al., 2005). The Tbx5 (LPM) that lies along the flank of the embryo. Embryological expressing cells form the inner core of the limb bud and are manipulations in the chick, together with genetic analysis in the enveloped by an epithelial layer of ectodermal cells. Tbx5 is mouse, have now provided a basic framework of the kinds of required not only for the initiation of the bud tissue interactions and gene activity that underlies the allocation primordia, but it also controls the outgrowth of the limb (for of cells to the limb primordia and their subsequent growth and example, see Rallis et al., 2003; Agarwal et al., 2003) — a differentiation (reviewed in Capdevila and Izpisua Belmonte, process that is dependant on the transfer of information from the 2001). Tbx5, a T-box containing transcriptional regulator, is an mesenchymal cells to the overlying ectoderm and the establish- evolutionarily conserved and a central determinant of the ment therein of a signaling center, the AER. forelimb bud developmental pathway. Expression of the Tbx5 A number of studies have now demonstrated that the gene is thought to be induced in mesenchymal cells of the LPM formation of the pectoral fins in the zebrafish is regulated by in response to signaling by the Fgf8 and Wnt2B that developmental processes that, in many ways, have been emanate from the adjacent (IM) (Cohn conserved during evolution. For instance, a homolog of tbx5 et al., 1995; Vogel et al., 1996; Crossley et al., 1996; Kawakami is expressed in an equivalent domain and acts in a similar manner in the determination of the fin primordia (Tamura et al., ⁎ Corresponding author. Fax: +65 6779 1117. 1999; Begemann and Ingham, 2000; Ruvinsky et al., 2000; Ahn E-mail address: [email protected] (S. Roy). et al., 2002; Garrity et al., 2002). Furthermore, the zone of

0012-1606/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2006.07.031 624 B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634 polarizing activity (ZPA), another important signaling source the tracheal system in the Drosophila embryo (de Souza et al., which secretes Sonic hedgehog (Shh), that polarizes the limb 1999; Wilm and Solnica-Krezel, 2005; Ng et al., 2006). bud along the antero-posterior axis, is also active in the Furthermore, targeted deletion of the Blimp-1 in the zebrafish fin bud (Neumann et al., 1999). Despite these and mouse results in embryonic lethality, with severe defects in the several other similarities, there are also notable differences in development of the branchial arches and a complete absence of the mechanism of morphogenesis of fins and tetrapod limbs. the primordial germ cells (Vincent et al., 2005; Ohinata et al., In the latter, AER formation is triggered by inductive 2005). In this report, we demonstrate that blimp-1 has an signaling mediated by Fgf10 that is secreted from the limb additional role in regulating the development of the pectoral bud mesenchyme. In response to Fgf10, the AER expresses fins in the zebrafish embryo. Here, its activity is critically two other Fgf signaling molecules, Fgf4 and Fgf8, which not required for the establishment of the ZPA and the AER through only maintain the expression of Fgf10 in the underlying the coordination of gene expression programs in the mesench- mesenchyme, but also direct the establishment of the ZPA. In ymal cells as well as the ectoderm. Consequently, in the mouse Fgf10 mutants, AER formation does not occur at all absence of Blimp-1 activity, specification of the fin primordia and Shh is never activated in the ZPA (Min et al., 1998; progresses normally, but subsequent events of fin outgrowth Sekine et al., 1999). Moreover, conditional inactivation of and patterning are completely arrested. Fgf8 function in the early limb ectoderm results in mice with substantial defects in limb formation (Lewandoski et al., Materials and methods 2000). By contrast, zebrafish fgf10 has a comparatively subservient role in fin development, functioning largely as a Zebrafish strains maintenance factor for the AER (Norton et al., 2005). As for The ubotp39 and the shh null mutant strain sonic you (syut4) were isolated in , it is a relatively late AER marker that does not appear to mutagenesis screens at the Max-Planck-Institut für Entwicklungsbiologie, have a vital role in the formation of the fin (Reifers et al., Tübingen (van Eeden et al., 1996; Schauerte et al., 1998). The strain carrying a 1998). These significant points of differences between complete loss-of-function allele of the zebrafish smoothened (smo) gene, slow- b641 zebrafish pectoral fin and amniote forelimb development muscle-omitted (smu ), was kindly provided by S. Devoto (Barresi et al., have been postulated to center, largely, on the involvement of 2000). an additional Fgf family member, Fgf24, early in the genetic Morpholino injections cascade controlling fin bud specification (Fischer et al., 2003). Fgf24 not only directs the expression of fgf10 in the The following antisense morpholinos (MOs) were used: blimp-1 splice site mesenchyme, but also is essential for establishing shh targeting MO and fgf24 and tbx5 MOs targeting their respective translational expression in the ZPA. Additionally, Fgf24 is an early marker start sites (Ahn et al., 2002; Fischer et al., 2003; Baxendale et al., 2004). The of the AER, making it a candidate signal responsible for the oligonucleotides were solubilized in sterile water and injected into newly fertilized zebrafish eggs at concentrations ranging from 5 to 10 ng/embryo. maintenance of fgf10 and shh expression in the mesenchyme, Efficacy of the blimp-1 splice MO was determined by RT-PCR, using primers a role that is analogous to Fgf8 of amniotes. that are complementary to sequences in exon 1 and exon 5 of the zebrafish Regardless of all of these advancements from investigations blimp-1 gene. The sequences of the primer pair are as follows: forward primer in different vertebrates, our understanding of many aspects of 5′-TCACTTACCATCTGGACTAGCA-3′, reverse primer 5′-CTTCGGT- ′ the pathway is far from complete. In TGCTTGCTGCTTG-3 . Sequencing of the amplified band obtained from the morphant embryos revealed the retention of the whole of intron 2 in their mis- particular, we do not fully understand how distinct patterns of spliced mRNA. gene expression are established in response to the variety of signals that have been recognized to function in the developing In situ hybridization and Alcian blue staining limb bud. During the differentiation of B-cells of the immune system, the Blimp-1 (for B-lymphocyte- Whole-mount in situ hybridization was performed following routine inducing maturation ) plays an important role in protocols. For colorimetric analyses, Digoxigenin (DIG) labeled antisense promoting their conversion into antibody secreting plasma RNA probes for the following were used: blimp-1 (Baxendale et al., 2004), tbx5 (Ruvinsky et al., 2000), fgf24 (Fischer et al., 2003), fgf10 (Ng et al., 2002), cells (Shapiro-Shelef and Calame, 2004). Our previous work shh (Krauss et al., 1993), dlx2a (Akimenko et al., 1994) and fgf8 (Reifers et al., with the zebrafish homolog of Blimp-1, U-boot (Ubo), has 1998). DIG antisense RNAs, together with those labeled with Fluorescein, were shown that its activity is also required for the specification of used for the simultaneous detection of blimp-1 and tbx5 and blimp-1 and dlx2a the slow-twitch muscle fibers in the myotome and the neural expression in the fin primordia. For these double fluorescent in situ hybridization crest progenitor cells at the boundary between the epidermis reactions, signals were developed using the Tyramide Signal Amplification (TSA) kit (Molecular Probes), according to the manufacturer's instructions. and the neural plate (Roy et al., 2001; Roy and Ng, 2004; Alcian blue staining of the fin endoskeleton was done as described previously Baxendale et al., 2004). blimp-1 expression is extremely (Grandel and Schulte-Merker, 1998). dynamic in embryos of all species examined, indicating that it has multiple functions in a variety of cell and tissue-types Image analysis and figure preparation during development (de Souza et al., 1999; Chang et al., 2002; Ha and Riddle, 2003; Baxendale et al., 2004; Wilm and Stained embryos were examined and photographed using a Zeiss compound microscope (Axioplan 2) equipped with a Nikon camera (DMX1200) for digital Solnica-Krezel, 2005; Vincent et al., 2005; Ng et al., 2006). In image capture. Optical sections of the fluorescent in situ hybridization stainings line with this, the gene has also been shown to regulate were obtained using a Zeiss LSM confocal microscope. Figures were assembled patterning of the gastrula in fish and frogs and differentiation of using Adobe Photoshop 6.01. B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634 625

Results time (Fig. 1D). This is in contrast to the tbx5 gene, whose expression is always restricted exclusively to the mesenchyme blimp-1 is dynamically expressed in the developing zebrafish (Fig. 1C). By 36 hpf, blimp-1 expression evolves into stronger pectoral fin bud levels and localizes to the apical fin fold — the zebrafish equivalent of the AER of amniote embryos (Fig. 1E). The During embryogenesis in the zebrafish, blimp-1 is expressed expression in the AER refines into a narrow fringe by 48 hpf, at in a variety of cells and tissues that include the developing the extreme edge of the outgrowing fin (Fig. 1F). At 72 hpf, pectoral fin anlagen, although the details of the pattern have not blimp-1 expression is almost completely extinguished from the been reported previously (Baxendale et al., 2004; Wilm and differentiating pectoral fin (Fig. 1G). Solnica-Krezel, 2005). We performed whole-mount in situ To more precisely visualize the spatial domains of this hybridization to fully document the temporal profile and the dynamic profile of blimp-1 expression, we performed double spatial domain of blimp-1 expression in the pectoral fin bud. label fluorescent in situ hybridization. At 24 hpf, we found Expression in this region is first detected very weakly at blimp-1 transcripts in the ectoderm and in the underlying approximately 20 h post fertilization (hpf); the levels gradually mesenchyme, where it colocalized with tbx5 (Figs. 2A–C). At intensify, such that by 24 hpf, prominent blimp-1 expression 30 hpf, however, the crescent-shaped pattern of blimp-1 can be observed in a distinct cluster of cells at the site of the superimposed almost entirely with dlx2a (Figs. 2D–F), a forming fin bud (Fig. 1A). Lateral views of stained embryos at zebrafish homologue of the distal-less family of this stage clearly reveal that the expression is localized to the genes that is an early marker of the fin ectoderm and is inner mesenchymal layer (Fig. 1B). blimp-1 expression also expressed in the AER of all vertebrates (Akimenko et al., 1994). appears to be present in the overlying ectodermal cells (Fig. These observations unequivocally confirm the notion that 1B). The ectodermal expression is even more apparent at blimp-1 is expressed in the fin bud mesenchyme as well as 30 hpf, when it assumes a crescent-shaped pattern, whereas the the ectoderm in a developmental stage-dependent manner. In expression in the mesenchyme dramatically decreases by this the early fin bud, blimp-1 in transcribed in the mesenchyme and

Fig. 1. Spatio-temporal profile of blimp-1 expression in the developing pectoral fin. (A) A wild-type embryo, showing blimp-1 expression in the fin bud (black arrows). Expression in the pharyngeal endoderm (white arrows) and spinal cord neurons (arrowheads) is also indicated. (B) Expression at this stage is evident in the ectoderm (black arrow) and the mesenchyme (white arrow). (C) tbx5 expression shows localization only to the mesenchyme (white arrow) and is excluded from the ectoderm (black arrow). (D) blimp-1 expression in the mesenchyme decreases and strengthens in the ectoderm (arrow). (E–G) blimp-1 expression in succeeding stages of embryogenesis. Panel A depicts dorsal view, all others depict lateral views. All panels of this and subsequent figures are oriented anterior to the left. 626 B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634

Fig. 2. Comparative analysis of the mesenchymal and ectodermal components of blimp-1 expression with respect to tbx5 and dlx2a using double label fluorescent in situ hybridization. (A) A wild-type embryo, showing blimp-1 expression in the fin bud. (B) The same embryo showing tbx5 expression in the fin bud. The ectoderm, which is devoid of tbx5 transcripts, is indicated (arrows). (C) Superimposition of the images depicted in panels A and B, showing colocalization of blimp-1 and tbx5 signals in the mesenchyme, while ectodermal cells only contain blimp-1 transcripts (arrows). (D–F) blimp-1 and dlx2a are co-expressed in cells of the AER from 30 hpf (arrow). All panels depict lateral views.

Fig. 3. Pectoral fin development is affected in embryos compromised in Blimp-1 activity. (A) Normarski image of a wild-type fin. (B) Fin of a homozygous ubo mutant embryo. (C) Rudimentary fin bud of a blimp-1 morphant embryo. (D–F) Alcian blue stain of the pectoral fin skeletal elements of a wild-type, ubo mutant and blimp-1 morphant embryo. 1=cleithrum, 2=scapulocoracoid, 3=endoskeletal disk, 4=actinotrichs, 5=postcoracoid process. All panels show lateral views. B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634 627 in the ectoderm, whereas during later period of fin outgrowth twitch muscle precursors as well as the progenitors of the neural and differentiation, its expression is limited to the ectodermal crest is less severe in ubo embryos and contrasts with the more cells of the AER. dramatic effects that are observed in these cells when expression of the Blimp-1 protein is “knocked down” using antisense MOs blimp-1 is required for the development of the pectoral fin against the blimp-1 gene (Roy and Ng, 2004; Baxendale et al., 2004). Indeed, RT-PCR analysis revealed the occurrence of The dynamic pattern of blimp-1 expression in the fin aberrant splicing of blimp-1 pre-mRNA in the morphants (i.e. mesenchymal cells as well as the ectoderm is suggestive of its embryos injected with splicing inhibitory anti-blimp-1 MOs), requirement for the specification and/or the proper outgrowth leading to the retention of the whole of intron 2 (Figs. 4A, B). If of the fin bud. To analyze this, we examined the pectoral fins translated, such a mis-spliced mRNA is predicted to produce a of embryos homozygous for a hypomorphic allele of the ubo severely truncated and non-functional Blimp-1 protein (Fig. locus that impairs wild-type activity of the Blimp-1 protein 4B). Consistent with this, the morphant embryos almost (Roy et al., 2001; Baxendale et al., 2004). At 72 hpf, these completely lack any visible signs of fin outgrowth; when embryos exhibit variably shortened pectoral fins with very examined at 72 hpf, these embryos exhibit a very small and irregular edges compared to that of their normal siblings, undifferentiated tubercular structure in the region where the indicating that Blimp-1 function is indeed necessary for pectoral fin is normally located (Fig. 3C; see also Wilm and proper development of the pectoral fin (Figs. 3A, B; see Table Solnica-Krezel, 2005). The internal skeleton of the wild-type fin 1 for a quantitative analysis of this and all other fin consists of a series of cartilaginous elements — the cleithrum, phenotypes described in this paper). scapulocoracoid and endoskeletal disk, arranged in that order We reasoned, however, that the hypomorphic nature of the along the proximo-distal axis, respectively (Fig. 3D) (Grandel ubo mutation is likely to mask a more critical function of and Schulte-Merker, 1998). Although fins of ubo mutants con- blimp-1 in the formation of the pectoral fin. We have previously tain all of these elements (Fig. 3E), albeit sometimes reduced in demonstrated that the mis-specification phenotype of the slow- size, the rudimentary fin buds of blimp-1 morphant embryos

Table 1 Quantitative analysis of the effects of the different mutants and morphants on gene expression in the developing fin bud and morphology of the differentiated pectoral fin Phenotype Genotype and number examined (n) Penetrance (and expressivity) Fin morphology at 72 hpf ubo=18 100% with fin truncation (extent of fin truncation was variable) blimp-1 morphants=35 100% with rudimentary fin bud Fin skeleton at 120 hpf ubo=8 37.5% showed shortened scapulocoracoid (shortening was variable) the rest appeared wild-type blimp-1 morphants=7 100% showed absence of elements distal to cleithrum tbx5 expression at 30 hpf blimp-1 morphants=12 100% showed wild-type like expression tbx5 expression at 36 hpf blimp-1 morphants=10 40% showed slight reduction in levels, the rest appeared wild-type tbx5 expression at 48 hpf blimp-1 morphants=11 63.6% showed reduction in levels, the rest appeared wild-type fgf24 expression at 24 hpf blimp-1 morphants=12 100% showed wild-type like expression fgf24 expression at 48 hpf blimp-1 morphants=14 100% showed absence of expression in ectoderm and sustained expression in mesenchyme blimp-1 expression at 24 hpf tbx5 morphants=21 100% showed complete absence from the fin bud region fgf24 morphants=12 100% showed complete absence from the fin bud region syu=15 100% showed wild-type like expression smu=15 100% showed wild-type like expression blimp-1 expression at 36 hpf syu=12 100% showed reduction in levels smu=10 100% showed reduction in levels blimp-1 expression at 48 hpf syu=8 100% showed reduction in levels smu=10 100% showed reduction in levels fgf10 expression at 26 hpf ubo=12 66.6% showed reduced expression, the rest appeared wild-type blimp-1 morphants=17 100% showed strong reduction fgf10 expression at 30 hpf ubo=19 20% showed reduced levels, the rest appeared wild-type blimp-1 morphants=16 100% showed strong reduction fgf10 expression at 36 hpf ubo=12 100% showed more or less wild-type levels of expression blimp-1 morphants=19 31.6% showed complete absence, the rest showed very strong reduction shh expression at 30 hpf ubo=11 100% showed reduced levels of expression blimp-1 morphants=18 100% showed complete absence of expression shh expression at 48 hpf ubo=9 100% showed reduced expression blimp-1 morphants=21 100% showed complete absence of expression dlx2a expression at 32 hpf ubo=18 100% showed reduced expression blimp-1 morphants=17 100% showed complete absence of expression dlx2a expression at 36 hpf ubo=13 100% showed reduced expression blimp-1 morphants=24 100% showed absence of expression fgf8 expression at 38 hpf ubo=18 66.7% embryos showed strong reduction in levels, the rest showed complete absence of expression blimp-1 morphants=15 100% showed absence of expression 628 B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634

Fig. 4. Splice junction targeted anti-blimp-1 MOs effectively block splicing of blimp-1 pre-mRNA. (A) RT-PCR of mRNA extracted from wild-type embryos and blimp-1 morphants, showing the expected 783 bp band in the wild-type lanes and an approximately 3 kb band in those of the morphants. The primer pair used for the PCR reaction amplifies across exons 1–5 (see Materials and methods). (B) Diagram illustrating the target site of the MO, at the junction between exon 2 and intron 2. The last codon of exon 2 and the first codon of exon 3, together with their corresponding amino acid, are indicated. The premature stop codon in the mis- spliced blimp-1 mRNA is highlighted in red. completely lack the skeletal structures distal to the cleithrum Ruvinsky et al., 2000). Consequently, we found that the (Fig. 3F). Based on all of these observations, we conclude that expression of tbx5 at early stages of fin primordia formation Blimp-1 plays a crucial role in the development of the pectoral occurs normally in ubo mutants as well as blimp-1 morphant fin. embryos, confirming that the activation of this gene in the fin mesenchyme is independent of Blimp-1 activity (Figs. 5A, B; blimp-1 acts downstream of tbx5 and fgf24 in the development data not shown). Later in development, although the pattern of of the fin primordium tbx5 in ubo mutants and their wild-type siblings appears indistinguishable, the levels are discernibly reduced and the In order to position blimp-1 in the genetic pathway that domain of expression smaller in the blimp-1 morphant embryos regulates the specification and patterning of the pectoral fin, we (Figs. 5C–F; data not shown). In this respect, our results first analyzed the expression of two important genes, tbx5 and contradict an earlier preliminary observation that had implicated fgf24, that act early in the induction of the fin primordium, in a role for blimp-1 upstream of tbx5 (Wilm and Solnica-Krezel, embryos compromised in Blimp-1 activity. A comparative 2005). study of the onset of expression of the three genes, i.e. tbx5, The expression of fgf24 in the fin mesenchymal cells of ubo fgf24 and blimp-1, suggests that blimp-1 is likely to function embryos and blimp-1 morphants appears identical to wild-type downstream from tbx5 as well as fgf24. While the earliest time embryos at 24 hpf (Figs. 5G, H; data not shown). Unlike tbx5, point of blimp-1 expression in the fin mesenchyme is 20 hpf, whose expression remains confined to the mesenchyme fgf24 is observed in this region from 18 hpf (Fischer et al., throughout fin development, fgf24 expression normally 2003) and tbx5 from 17 hpf (Begemann and Ingham, 2000; declines in these cells between 28 and 30 hpf and reappears B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634 629

Fig. 5. blimp-1 acts downstream of tbx5 and fgf24.(A–F) tbx5 expression in the fin buds (arrows) of wild-type and blimp-1 morphants. (G and H) fgf24 expression (arrows) in a wild-type embryo and a blimp-1 morphant. (I) A wild-type fin, showing fgf24 expression in the AER (arrow). (J) A blimp-1 morphant at the same stage, showing absence of fgf24 from the ectoderm (black arrow) and its continued expression in the mesenchyme (white arrow). (K and L) blimp-1 expression is absent (arrows) from the prospective fin buds of tbx5 and fgf24 MO injected embryos. All panels depict dorsal views, except panels I and J, which depict lateral views. in the AER (Fischer et al., 2003). In line with this, when wild- be first detected in the mesenchymal cells at 24 hpf (Ng et al., type embryos were examined at 48 hpf, we found prominent 2002; Fischer et al., 2003). Moreover, loss of fgf24 expression of fgf24 in the AER and complete absence from the completely inhibits fgf10 expression (Fischer et al., 2003). fin mesenchyme (Fig. 5I). A similar pattern of fgf24 was Since blimp-1 expression also requires Fgf24 activity, this observed in ubo mutant embryos, although some of them would indicate that Blimp-1 could be needed for inducing the showed a lower level of expression than that apparent among expression of fgf10 in the mesenchyme, in response to Fgf24 their wild-type siblings (data not shown). Strikingly, in blimp-1 signaling. In line with such a possibility, we found that, at morphants, fgf24 sustains its expression in the mesenchymal 26 hpf, the levels of fgf10 transcripts are reduced in the fin buds cells and fails to get activated in the fin bud ectoderm (Fig. 5J). of ubo mutant embryos compared to their wild-type siblings We have also made the reciprocal analysis of the status of (Figs. 6A, B), although the expression levels appeared more or blimp-1 transcription in the fin primordia of embryos that are less comparable later, at 30 and 36 hpf (Figs. 6D, E, G, H). In depleted of the Tbx5 and Fgf24 proteins. We failed to observe blimp-1 morphants, which represent a much stronger loss-of- any blimp-1 expression in tbx5 and fgf24 morphants in the function condition, there was a considerable reduction of fgf10 region of the prospective fin bud at all stages of embryogenesis expression in the mesenchymal cells at all of these stages of (Figs. 5K, L; data not shown). These data confirm the view that development (Figs. 6C, F, I). Thus, a primary defect in the fin blimp-1 operates downstream of tbx5 and fgf24 in the early buds of embryos lacking Blimp-1 activity is their inability to genetic cascade that specifies the pectoral fin bud. properly institute the expression of fgf10 in the mesenchyme.

Absence of Blimp-1 function prevents proper induction of fgf10 Blimp-1 induces shh in the ZPA and requires Hh signaling for in the mesenchyme the maintenance of its own expression

We next investigated the effects of the loss of Blimp-1 on In amniotes, induction of Shh expression and formation of the expression of fgf10 in the fin bud mesenchyme. During the ZPA in the posterior mesenchymal cells are directed by normal development, fgf10 expression follows fgf24 and can the Fgf proteins secreted from the AER (Sun et al., 2002; 630 B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634

Fig. 6. Blimp-1 activity is required for fgf10 expression in the fin bud mesenchyme. (A–I) fgf10 expression in the fin mesenchyme (arrows) of wild-type, ubo and blimp-1 morphant embryos at specific developmental stages. All panels depict dorsal views.

Boulet et al., 2004). Instead, in the zebrafish embryo, shh detected in the ZPA around 28 hpf (Krauss et al., 1993). This expression is initiated differently, by Fgf24 signaling, derived would implicate that, in the context of the fin, Hh signaling is from the mesenchyme (Fischer et al., 2003). Here, the AER not required for the induction of blimp-1. Indeed, embryos seems to be only involved in the maintenance of shh lacking activity of Shh or Smoothened (Smo), a transmem- expression in the ZPA. Since Blimp-1 functions downstream brane protein that is essential of the intracellular transduction of fgf24 and is required for the full activation of fgf10 of the Hh signal, showed normal levels and pattern of blimp-1 expression, we explored whether embryos lacking blimp-1 transcription in the fin primordia at 24 hpf (Figs. 7G, J). function also show a loss of shh expression from the ZPA. However, Hh activity does play a role in the maintenance of shh is expressed at lower levels in ubo embryos compared blimp-1 expression through the succeeding stages of fin to their wild-type counterparts at all developmental stages development as evidenced by the progressive decline in analyzed, indicating that activation of the shh gene requires a blimp-1 transcription in the absence of Hh pathway activity threshold level of Blimp-1 activity that is reduced in the ubo (Figs. 7H–L). embryos (Figs. 7A–E). Consistent with this, in the blimp-1 morphants, shh expression is never observed in the pectoral Blimp-1 activity is essential for the specification of the AER fin primordia (Figs. 7C, F). Thus, Blimp-1 is also required for the activation of shh expression and the establishment of Since Fgf10 signaling relays inductive information from the ZPA in the posterior mesenchymal cells of the developing the mesenchyme to the ectoderm and its activity is necessary fin bud. for the proper development of the AER, we reasoned that loss We also examined the reciprocal consequence: that of the of Blimp-1 function, which affects fgf10 expression, should loss of Hh signaling on blimp-1 expression in the fin bud. We also affect the AER. More importantly, blimp-1 is itself have previously shown that blimp-1 is activated in the actively transcribed in the ectoderm and in the AER from precursors of the slow-twitch muscles within the of early stages of fin development, signifying that it could the zebrafish embryo in response to Hh signaling that directly influence the expression of marker genes in the emanates from midline tissues (Baxendale et al., 2004). It is ectoderm and, consequently, the formation of the AER. The apparent from data presented here that, in the pectoral fin fact that ectodermal gene expression indeed does get affected buds, blimp-1 expression precedes the onset of shh in the in the absence of Blimp-1 function is already borne out from ZPA. Whereas blimp-1 initiates as early as 20 hpf, shh is first our earlier observation that fgf24 fails to get activated in the B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634 631

Fig. 7. Blimp-1 function is necessary for shh expression whereas Hh signaling is required for maintenance, but not the initiation, of blimp-1 expression. (A–F) shh expression (arrows) in the ZPA of wild-type, ubo mutant and a blimp-1 morphant embryos at specific developmental stages. (G–I) blimp-1 expression (arrows) in embryos lacking Shh activity. (J–L) blimp-1 expression (arrows) in embryos lacking Smo activity. Panels A–F, G and J depict dorsal views; panels H, I, K and L show lateral views. ectoderm in blimp-1 morphant embryos (Fig. 5J). Analysis of Discussion the expression of dlx2a showed that it is quite noticeably reduced in the fin buds of the ubo mutants and is totally We have identified that the transcription factor Blimp-1 is a undetectable in the blimp-1 morphant embryos (Figs. 8A–F). novel component of the regulatory pathway that directs the The fgf8 gene, which is the definitive marker of the AER, is development of the pectoral fin in the zebrafish embryo. The activated in the zebrafish fin bud much later compared to that in spatio-temporal expression of blimp-1 in the fin bud mesen- birds and mammals (Reifers et al., 1998). In wild-type embryos, chyme and epistasis analysis allowed us to position the gene fgf8 is first detectable in cells of the AER around 38 hpf, at downstream of tbx5 and fgf24, but upstream of fgf10 in the fin about the time the AER becomes morphologically distinguish- development pathway. Accordingly, we have shown that loss of able (Fig. 8G). fgf8 expression is barely visible in the fin buds Blimp-1 interrupts fin development at an early stage, immedi- of ubo mutants and is completely absent from those of the ately following the establishment of the fin primordia, blimp-1 morphants (Figs. 8H, I). These dramatic effects on precluding proper initiation of fgf10 expression in the fin bud dlx2a and fgf8 expression underscore a pivotal role for Blimp-1 mesenchyme and the establishment of the ZPA. As a con- in the specification of the AER. sequence, ectoderm and AER markers genes like dlx2a, fgf24 632 B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634

Fig. 8. Blimp-1 activity is crucial for the establishment the AER. (A–F) dlx2a expression in the AER (arrow) of a wild-type, ubo mutant and a blimp-1 morphant embryo. (G–I) fgf8 expression in the AER (arrow) of a wild-type, ubo mutant and a blimp-1 morphant embryo. All panels depict lateral views of developing pectoral fin buds. and fgf8 that depend on Fgf10 signaling are completely down- feedback loop of gene expression that is necessary for the regulated. However, our observation that blimp-1 is also one of outgrowth and differentiation of the fin. Veracity of this model the earliest markers of the fin ectoderm and, subsequently, the comes from the observation of fin development in the recently AER itself, denotes that the Blimp-1 protein has an independent described fgf10 mutants that are devoid of the Fgf10 protein role in activating gene expression here that is distinct from its (Norton et al., 2005). Here, in contrast to Blimp-1 deficient function in the mesenchyme. Evidence for such a scenario is embryos, the early phase of gene expression in the fin ectoderm derived from the examination of AER development in embryos and the initiation of the AER and the ZPA occur almost homozygous for the hypomorphic ubo mutation that reduces, normally. However, all of these fail to be maintained in the but does not completely eliminate, the activity of the Blimp-1 succeeding stages of embryogenesis. Ultimately, they are totally protein. In these animals, specification of the fin bud and lost, and fin outgrowth is completely inhibited. On similar lines, induction of fgf10 progresses almost normally, but the we have shown that, although the initiation of blimp-1 expression of AER-specific genes such as dlx2a and fgf8 are expression in the fin primordia precedes the onset of shh in appreciably reduced and distal elements of the mature fin are the ZPA and is regulated independently of Shh activity, Hh truncated. signaling is nevertheless required for the continued expression To integrate all of our findings, we propose that Fgf24 of blimp-1 through the later stages of fin outgrowth. signaling is responsible for the activation of the blimp-1 gene in We note that the homologs of blimp-1 have previously been the mesenchyme as well as in the ectodermal cells of the fin observed to be expressed in both fore- and the hindlimb buds bud. The Blimp-1 protein then participates, directly or of the developing chick and the mouse embryo (Chang et al., indirectly, in the induction of the expression of fgf10 and shh 2002; Ha and Riddle, 2003; Vincent et al., 2005). In the chick, in the mesenchymal cells, as well as genes such as dlx2a and expression is always restricted to the ectoderm — it originates fgf24 in the ectoderm of the early fin bud. Further progression in the dorsal ectoderm and then becomes prominent in the of fin development then becomes dependant on signaling by AER. In the mouse, Blimp-1 transcripts are first present Fgf10 from the mesenchyme. In this model, maintenance of throughout the limb buds and then shift to the posterior region blimp-1 expression in the AER by Fgf10 will result in the that includes the ZPA. Expression is also prevalent in the AER. maintenance of fgf24 in this tissue, and subsequently, in the These species-specific disparities in the expression pattern induction of fgf8 and . Secretion of all of these Fgf proteins parallel the differences that are evident in the requirement of from the AER will, in turn, ensure the continued expression of the gene in the limb development program of different animals. fgf10 in the mesenchyme and shh in the ZPA, eliciting the For example, Blimp-1 activity is not only dispensable for the B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634 633 initial events of limb bud specification in the mouse, but de Souza, F.S., Gawantka, V., Gomez, A.P., Delius, H., Ang, S.L., Niehrs, C., unlike in the zebrafish, it also appears not to be necessary for 1999. The gene Xblimp1 controls anterior endomesodermal cell fate in Spemann's organizer. EMBO J. 18, 6062–6072. the formation of the ZPA and the AER (Vincent et al., 2005). Fischer, S., Draper, B.W., Neumann, C.J., 2003. The zebrafish fgf24 mutant However, the possibility does remain that Blimp-1 has an identifies an additional level of Fgf signaling involved in vertebrate forelimb important later role in patterning mammalian limbs that is initiation. Development 130, 3515–3524. obscured by the premature lethality of the mutant embryos. Garrity, D.M., Childs, S., Fishman, M.C., 2002. The heartstrings mutation in Ablation of its activity specifically in the limb precursor cells zebrafish causes heart/fin Tbx5 deficiency syndrome. Development 129, 4635–4645. should help to fully clarify this issue. Nevertheless, Grandel, H., Schulte-Merker, S., 1998. The development of the paired fins in the irrespective of what the precise role of Blimp-1 might be in zebrafish (Danio rerio). Mech. Dev. 79, 99–120. the limb buds of amniotes, the lack of an early phenotype in Ha, A.S., Riddle, R.D., 2003. cBlimp-1 expression in chick limb bud the mouse helps to reinforce the idea that alterations in the development. Gene Expr. Patterns 3, 297–300. regulation and function of genes and their networks are likely Kawakami, Y., Capdevila, J., Buscher, D., Itoh, T., Rodriguez, E.C., Izpisua Belmonte, J.C., 2001. WNT signals control FGF-dependent limb initiation to be the developmental basis for the morphological and AER induction in the chick embryo. Cell 104, 891–900. diversification of appendages apparent in the different groups Krauss, S., Concordet, J.P., Ingham, P.W., 1993. A functionally conserved of vertebrates. homolog of the Drosophila segment polarity gene hh is expressed in tissues with polarizing activity in zebrafish embryos. Cell 75, 1431–1444. Acknowledgments Lewandoski, M., Sun, X., Martin, G.R., 2000. Fgf8 signaling from the AER is essential for normal limb development. Nat. Genet. 26, 460–463. Min, H., Danilenko, D.M., Scully, S.A., Bolon, B., Ring, B.D., Tarpley, J.E., We are grateful to G. Begemann, R. Ho, M. Westerfield, C. DeRose, M., Simonet, W.S., 1998. Fgf-10 is required for both limb and lung Neumann and B. Draper for kindly providing reagents and J. Xu development and exhibits striking functional similarity to Drosophila for his advice and assistance in the molecular analysis of the branchless. Genes Dev. 12, 3156–3161. splicing defects in the blimp-1 morphants. This work was Neumann, C.J., Grandel, H., Gaffield, W., Schulte-Merker, S., Nusslein-Volhard, C., 1999. Transient establishment of anteroposterior polarity in the zebrafish funded by the Institute of Molecular and Cell Biology and the pectoral fin bud in the absence of sonic hedgehog activity. Development 126, Agency for Science, Technology and Research of Singapore. 4817–4826. S.R. is an adjunct faculty of the Department of Biological Ng, J.K., Kawakami, Y., Buscher, D., Raya, A., Itoh, T., Koth, C.M., Rodriguez, Sciences, National University of Singapore. E.C., Rodriguez-Leon, J., Garrity, D.M., Fishman, M.C., Izpisua Belmonte, J.C., 2002. The limb identity gene Tbx5 promotes limb initiation by interacting with Wnt2b and fgf10. Development 129, 5161–5170. References Ng, T., Yu, F., Roy, S., 2006. A homologue of the vertebrate SET domain and zinc finger protein Blimp-1 regulates terminal differentiation of the tracheal Agarwal, P., Wylie, J.N., Galceran, J., Arkhitko, O., Li, C., Deng, C., Grosschedl, system in the Drosophila embryo. Dev. Genes Evol. 216, 243–252. R., Bruneau, B.G., 2003. Tbx5 is essential for forelimb bud initiation Norton, W.H., Ledin, J., Grandel, H., Neumann, C.J., 2005. HSPG synthesis by following patterning of the limb field in the mouse embryo. Development zebrafish Ext2 and Extl3 is required for Fgf10 signaling during limb 130, 623–633. development. Development 132, 4963–4973. Ahn, D.G., Kourakis, M.J., Rohde, L.A., Silver, L.M., Ho, R.K., 2002. T-box Ohinata, Y., Payer, B., O'Carroll, D., Ancelin, K., Ono, Y., Sano, M., Barton, gene tbx5 is essential for formation of the pectoral limb bud. Nature 417, S.C., Obukhanych, T., Nussenzweig, M., Tarakhovsky, A., Saitou, M., 754–758. Surani, M.A., 2005. Blimp1 is a critical determinant of the germ cell Akimenko, M.A., Ekker, M., Wegner, J., Lin, W., Westerfield, M., 1994. lineage in mice. Nature 436, 207–213. Combinatorial expression of three zebrafish genes related to distal-less: part Perantoni, A.O., Timofeeva, O., Naillat, F., Richman, C., Pajni-Underwood, S., of a homeobox gene code for the head. J. Neurosci. 14, 3475–3486. Wilson, C., Vainio, S., Dove, L.F., Lewandoski, M., 2005. Inactivation of Barresi, M.J., Stickney, H.L., Devoto, S.H., 2000. The zebrafish slow-muscle- Fgf8 in early mesoderm reveals an essential role in kidney development. omitted gene product is required for Hedgehog signal transduction and the Development 132, 3859–3871. development of slow muscle identity. Development 127, 2189–2199. Rallis, C., Bruneau, B.G., Del Buono, J., Seidman, C.E., Seidman, J.G., Nissim, Baxendale, S., Davison, C., Muxworthy, C., Wolff, C., Ingham, P.W., Roy, S., S., Tabin, C.J., Logan, M.P., 2003. Tbx5 is required for forelimb bud 2004. The B-cell maturation factor Blimp-1 specifies vertebrate slow-twitch formation and continued outgrowth. Development 130, 2741–2751. muscle fiber identity in response to Hedgehog signaling. Nat. Genet. 36, Reifers, F., Bohli, H., Walsh, E.C., Crossley, P.H., Stainier, D.Y., Brand, M., 88–93. 1998. Fgf8 is mutated in zebrafish acerebellar (ace) mutants and is Begemann, G., Ingham, P.W., 2000. Developmental regulation of Tbx5 in required for maintenance of midbrain–hindbrain boundary development zebrafish embryogenesis. Mech. Dev. 90, 299–304. and somitogenesis. Development 125, 2381–2395. Boulet, A.M., Moon, A.M., Arenkiel, B.R., Capecchi, M.R., 2004. The roles Roy, S., Ng, T., 2004. Blimp-1 specifies neural crest and sensory neuron of Fgf4 and fgf8 in limb bud initiation and outgrowth. Dev. Biol. 273, progenitors in the zebrafish embryo. Curr. Biol. 14, 1772–1777. 361–372. Roy, S., Wolff, C., Ingham, P.W., 2001. The u-boot mutation identifies a Capdevila, J., Izpisua Belmonte, J.C., 2001. Patterning mechanisms controlling Hedgehog-regulated myogenic switch for fiber-type diversification in the vertebrate limb development. Annu. Rev. Cell Dev. Biol. 17, 87–132. zebrafish embryo. Genes Dev. 15, 1563–1576. Chang, D.H., Cattoretti, G., Calame, K.L., 2002. The dynamic expression Ruvinsky, I., Oates, A.C., Silver, L.M., Ho, R.K., 2000. The evolution of paired pattern of B lymphocyte induced maturation protein-1 (Blimp-1) during appendages in vertebrates: T-box genes in the zebrafish. Dev. Genes Evol. mouse embryonic development. Mech. Dev. 117, 305–309. 210, 82–91. Cohn, M.J., Izpisua-Belmonte, J.C., Abud, H., Heath, J.K., Tickle, C., 1995. Schauerte, H.E., van Eeden, F.J., Fricke, C., Odenthal, J., Strahle, U., Haffter, P., Fibroblast growth factors induce additional limb development from the flank 1998. Sonic hedgehog is not required for the induction of medial floor plate of chick embryos. Cell 80, 739–746. cells in the zebrafish. Development 125, 2983–2993. Crossley, P.H., Minowada, G., MacArthur, C.A., Martin, G.R., 1996. Roles for Sekine, K., Ohuchi, H., Fujiwara, M., Yamasaki, M., Yoshizawa, T., Sato, T., fgf8 in the induction, initiation, and maintenance of chick limb develop- Yagishita, N., Matsui, D., Koga, Y., Itoh, N., Kato, S., 1999. Fgf10 is ment. Cell 84, 127–136. essential for limb and lung formation. Nat. Genet. 21, 138–141. 634 B.C. Lee, S. Roy / Developmental Biology 300 (2006) 623–634

Shapiro-Shelef, M., Calame, K., 2004. Plasma cell differentiation and multiple Vincent, S.D., Dunn, N.R., Sciammas, R., Shapiro-Shalef, M., Davis, M.M., myeloma. Curr. Opin. Immunol. 16, 226–234. Calame,K.,Bikoff,E.K.,Robertson,E.J.,2005.Thezincfinger Sun, X., Mariani, F.V., Martin, G.R., 2002. Functions of FGF signaling from the transcriptional repressor Blimp1/Prdm1 is dispensable for early axis apical ectodermal ridge in limb development. Nature 418, 501–508. formation but is required for specification of primordial germ cells in the Tamura, K., Yonei-Tamura, S., Belmonte, J.C., 1999. Differential expression of mouse. Development 132, 1315–1325. Tbx4 and Tbx5 in zebrafish fin buds. Mech. Dev. 87, 181–184. Vogel, A., Rodriguez, C., Izpisua-Belmonte, J.C., 1996. Involvement of FGF-8 van Eeden, F.J., Granato, M., Schach, U., Brand, M., Furutani-Seiki, M., Haffter, P., in initiation, outgrowth and patterning of the vertebrate limb. Development Hammerschmidt, M., Heisenberg, C.P., Jiang, Y.J., Kane, D.A., Kelsh, R.N., 122, 1737–1750. Mullins, M.C., Odenthal, J., Warga, R.M., Allende, M.L., Weinberg, E.S., Wilm, T.P., Solnica-Krezel, L., 2005. Essential roles of a zebrafish / Nusslein-Volhard, C., 1996. Mutations affecting formation and pattern- blimp1 homolog in embryo patterning and organogenesis. Development ing in the zebrafish, Danio rerio. Development 123, 153–164. 132, 393–404.